Quantification of litter in cities using a smartphone application and citizen science in conjunction with deep learning-based image processing DOI Creative Commons
Shin’ichiro Kako,

Ryunosuke Muroya,

Daisuke Matsuoka

et al.

Waste Management, Journal Year: 2024, Volume and Issue: 186, P. 271 - 279

Published: June 28, 2024

Cities are a major source of litter pollution. Determination the abundance and composition plastic in cities is imperative for effective pollution management, environmental protection, sustainable urban development. Therefore, here, multidisciplinary approach to quantify classify environments proposed. In present study, data collection was integrated via Pirika smartphone application conducted image analysis based on deep learning. launched May 2018 and, date, has collected approximately one million images. Visual classification revealed that most common types were cans, bags, bottles, cigarette butts, boxes, sanitary masks, order. The top six categories accounted 80 % total, whereas three more than 60 total imaged litter. A deep-learning processing algorithm developed automatically identify categories. Both precision recall derived from model higher 75 %, enabling proper categorization. quantity automated also plotted map using location acquired concurrently with images by application. Conclusively, this study demonstrates citizen science supported applications learning-based can enable visualization, quantification, characterization street cities.

Language: Английский

How much does marine litter weigh? A literature review to improve monitoring, support modelling and optimize clean-up activities DOI Creative Commons
Umberto Andriolo, Gil Gonçalves

Environmental Pollution, Journal Year: 2024, Volume and Issue: 361, P. 124863 - 124863

Published: Aug. 30, 2024

The weight of marine litter has been marginally considered in comparison to counting and categorizing items. However, determines dynamics on water coasts, it is an essential parameter for planning optimizing clean-up activities. This work reviewed 80 publications that reported both the number beached macro-litter worldwide. On average, a item weighed 19.5 ± 20.3 g, with median 13.4 g. Plastics composed 80% by 51% global bulk. A plastic 12.9 13.8 g 9 analysis based continents bodies returned similar values, which can be used estimate beaches from past future visual census surveys, remote sensing imagery. Overall, this improve monitoring reports support modelling, thereby contributing environmental protection mitigation efforts.

Language: Английский

Citations

5

Monitoring macroplastics in aquatic and terrestrial ecosystems: Expert survey reveals visual and drone-based census as most effective techniques DOI Creative Commons
Luca Gallitelli, Pierre Girard, Umberto Andriolo

et al.

The Science of The Total Environment, Journal Year: 2024, Volume and Issue: 955, P. 176528 - 176528

Published: Sept. 25, 2024

Language: Английский

Citations

5

Use of Drone Remote Sensing to Identify Increased Marine Macro-Litter Contamination following the Reopening of Salgar Beach (Colombian Caribbean) during Pandemic Restrictions DOI Open Access
Rogério Portantiolo Manzolli, Luana Portz

Sustainability, Journal Year: 2024, Volume and Issue: 16(13), P. 5399 - 5399

Published: June 25, 2024

This study involves an integrated and innovative approach employing high-frequency monitoring, which is rare in studies focusing on solid waste beaches. Eight drone flights were performed over a tourist beach the Colombian Caribbean to achieve two main objectives: (i) quantify changes marine macro-litter (>2.5 cm) density, differences between period when was closed due COVID-19 pandemic subsequent reopening period; (ii) map abundance of coast, with emphasis single-use waste. The number items litter increased 9-fold periods, found crisp/sweet packets (n = 304, 13% total waste), plastic cups 248, 11%), expanded polystyrene (food containers) 227, 10%). factors contributing presence distribution tourists, use beach, offshore wind direction. results revealed that Salgar Beach can be considered exporter since incorporated into longshore current redistributed either nearby beaches or ocean. emphasizes potential for using images monitoring as well efficiency programs combatting at sea.

Language: Английский

Citations

4

Developing a harmonized decision framework for shoreline marine debris monitoring across APEC economies DOI
Jongsu Lee, Jong-Myoung Lee, Sunwook Hong

et al.

Marine Pollution Bulletin, Journal Year: 2025, Volume and Issue: 215, P. 117877 - 117877

Published: April 3, 2025

Language: Английский

Citations

0

Spatiotemporal variation in marine litter distribution along the Bulgarian Black Sea sandy beaches: amount, composition, plastic pollution, and cleanliness evaluation DOI Creative Commons
Radoslava Bekova, Bogdan Prodanov

Frontiers in Marine Science, Journal Year: 2024, Volume and Issue: 11

Published: June 18, 2024

The threat of anthropogenic marine litter, particularly plastic pollution, to ecosystems and human health, has spurred mitigation initiatives global scientific research. Following the Marine Strategy Framework Directive guidelines, this study evaluated litter distribution, cleanliness, pollution indices along Bulgarian Black Sea beaches in 2023. survey integrates visual assessment, manual sampling, drone mapping, distributing coastline encompass a broader range, totaling 45, including 28 remote/natural, 10 semi-urban, 7 urban beaches. Results indicate 48% decrease distribution on from 1462 ± 147 items/100 m 2021 753 97 2023, with Artificial polymer materials/plastic materials constituting 88.62% total amount. A comprehensive macro assessment was carried out using PAI for first time. In average cleanliness status classified as “ moderate ” (CCI: 7.61 1.00), clean northern central contrasting dirty southern Urbanized were assessed highest level (PAI AV,23 : 5.51; CCI 18.16). long term, maintain values AV,18-23 8.81 0.89, 2.35 0.32, persisting throughout period, necessitating ongoing monitoring intervention strategies. Despite identifying significant number beaches, none meet EU threshold value 20 m. This highlights urgent need effective interventions combat accumulation or semi-urban emphasizing multi-stakeholder collaboration sustainable solutions coastal ecosystem preservation.

Language: Английский

Citations

2

Quantification of litter in cities using a smartphone application and citizen science in conjunction with deep learning-based image processing DOI Creative Commons
Shin’ichiro Kako,

Ryunosuke Muroya,

Daisuke Matsuoka

et al.

Waste Management, Journal Year: 2024, Volume and Issue: 186, P. 271 - 279

Published: June 28, 2024

Cities are a major source of litter pollution. Determination the abundance and composition plastic in cities is imperative for effective pollution management, environmental protection, sustainable urban development. Therefore, here, multidisciplinary approach to quantify classify environments proposed. In present study, data collection was integrated via Pirika smartphone application conducted image analysis based on deep learning. launched May 2018 and, date, has collected approximately one million images. Visual classification revealed that most common types were cans, bags, bottles, cigarette butts, boxes, sanitary masks, order. The top six categories accounted 80 % total, whereas three more than 60 total imaged litter. A deep-learning processing algorithm developed automatically identify categories. Both precision recall derived from model higher 75 %, enabling proper categorization. quantity automated also plotted map using location acquired concurrently with images by application. Conclusively, this study demonstrates citizen science supported applications learning-based can enable visualization, quantification, characterization street cities.

Language: Английский

Citations

1